Project 1 is a corollary biomarker evaluation we propose as a study nested in the ongoing ECOG 1697 intergroup trial. E1697 evaluates adjuvant high dose IFN-a2b given for 4 weeks IV 5/7 days a week vs. observation in patients with resected melanoma of intermediate risk: (1) T2b NO, (2) T3a-b NO, (3) T4a-b NO, (4) T1 -4 N1 -2a, (microscopic). Recent evidence suggests an association between autoimmunity and favorable melanoma outcome after high-dose IFN-a2b. We propose to first evaluate the presence of autoantibodies to multiple autoantigens of normal tissue as well as melanoma, utilizing ELISA or Luminex analysis of serial blood samples from 300 patients at baseline, 1, 3, 6, 9 and 12 months on IFN or observation, in E1697. This evaluation of autoimmunity as a marker of outcome of adjuvant IFN-a 2b will determine the predictive value of autoantibody responses individually and collectively. Patients may be genetically predisposed to developing autoimmunity with IFN-a2b therapy, and evidence supports specific HLA genotypes and polymorphic variations in the CTLA-4 and FOXP3 genes as potential immunogenetic markers that may predict the capacity for autoimmunity, and therefore, benefit from therapy with IFN-a2b. We will therefore determine whether patients are predisposed to autoimmunity induced by IFN-a 2b based on these immunogenetic markers in PBMC using PCR-based assays, and determine whether a genetic signature of adequately high sensitivity and specificity may predict development of autoimmunity induced by IFN-a2b treatment. We hypothesize that the analysis of the immune responses of patients who have developed autoimmunity, in the context of their MHC class l/ll composition, will allow the definition of new and more relevant antigens for future vaccine immunotherapy. In the context of the proposed studies we will measure current candidate serum markers including S-100 , LDH, CRP and MIA levels by ELISA, as well as a panel of cytokines by multiplex analysis for comparison to the autoimmune panel we have focused upon in this project. This study will evaluate immunologic, genetic, and other biomarkers to classify melanoma patients into groups with different prognosis and therapeutic potential. We will attempt to develop prediction models to identify groups of patients that are more or less likely to benefit from adjuvant therapy.